mlsanityData to 7 Oct 2026

Subgraph Pooling: Tackling Negative Transfer on Graphs

arXiv 2402.08907 ↗First posted 14 Feb 2024cs.LG cs.AI cs.SIcited by 17

Not on the map. Why?

Abstract (arXiv)

Transfer learning aims to enhance performance on a target task by using knowledge from related tasks. However, when the source and target tasks are not closely aligned, it can lead to reduced performance, known as negative transfer. Unlike in image or text data, we find that negative transfer could commonly occur in graph-structured data, even when source and target graphs have semantic similarities. Specifically, we identify that structural differences significantly amplify the dissimilarities in the node embeddings across graphs. To mitigate this, we bring a new insight in this paper: for semantically similar graphs, although structural differences lead to significant distribution shift in node embeddings, their impact on subgraph embeddings could be marginal. Building on this insight, we introduce Subgraph Pooling (SP) by aggregating nodes sampled from a k-hop neighborhood and Subgraph Pooling++ (SP++) by a random walk, to mitigate the impact of graph structural differences on knowledge transfer. We theoretically analyze the role of SP in reducing graph discrepancy and conduct extensive experiments to evaluate its superiority under various settings. The proposed SP methods are effective yet elegant, which can be easily applied on top of any backbone Graph Neural Networks (GNNs). Our code and data are available at: https://github.com/Zehong-Wang/Subgraph-Pooling.

Key facts. Cited by 17 papers in the map (data through 7 Oct 2026). Builds on 47 of its 59 references that are in the map. First posted on arXiv 14 Feb 2024.

Builds on

47 of 59 references.

References · 47

  1. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,282
  2. 2017Graph Attention Networkscited by 8,854
  3. 2017Inductive Representation Learning on Large Graphscited by 6,284
  4. 2015Domain-Adversarial Training of Neural Networkscited by 3,023
  5. 2019Fast Graph Representation Learning with PyTorch Geometriccited by 2,414
  6. 2020Open Graph Benchmark: Datasets for Machine Learning on Graphscited by 2,148
  7. 2019A Comprehensive Survey on Transfer Learningcited by 1,154
  8. 2020LightGCN: Simplifying and Powering Graph Convolution Network for Recommendationcited by 1,134
  9. 2019Multi-scale Attributed Node Embeddingcited by 543
  10. 2020GCC: Graph Contrastive Coding for Graph Neural Network Pre-Trainingcited by 461
  11. 2019Self-Attention Graph Poolingcited by 421
  12. 2019PairNorm: Tackling Oversmoothing in GNNscited by 297
35 more
  1. 2018Characterizing and Avoiding Negative Transfercited by 226
  2. 2017struc2vec: Learning Node Representations from Structural Identitycited by 214
  3. 2017Central Moment Discrepancy (CMD) for Domain-Invariant Representation Learningcited by 192
  4. 2022Discovering Invariant Rationales for Graph Neural Networkscited by 186
  5. 2022Handling Distribution Shifts on Graphs: An Invariance Perspectivecited by 163
  6. 2019Anti-Money Laundering in Bitcoin: Experimenting with Graph Convolutional Networks for Financial Forensicscited by 158
  7. 2023All in One: Multi-task Prompting for Graph Neural Networkscited by 147
  8. 2022G-Mixup: Graph Data Augmentation for Graph Classificationcited by 145
  9. 2022Learning Causally Invariant Representations for Out-of-Distribution Generalization on Graphscited by 144
  10. 2022GOOD: A Graph Out-of-Distribution Benchmarkcited by 140
  11. 2020A Survey on Negative Transfercited by 120
  12. 2020Graphon Neural Networks and the Transferability of Graph Neural Networkscited by 117
  13. 2022Not too little, not too much: a theoretical analysis of graph (over)smoothingcited by 106
  14. 2020Transfer Learning of Graph Neural Networks with Ego-graph Information Maximizationcited by 93
  15. 2019Transferability of Spectral Graph Convolutional Neural Networkscited by 90
  16. 2021Shift-Robust GNNs: Overcoming the Limitations of Localized Graph Training Datacited by 86
  17. 2019Graph Transfer Learning via Adversarial Domain Adaptation with Graph Convolutioncited by 67
  18. 2023Energy-based Out-of-Distribution Detection for Graph Neural Networkscited by 67
  19. 2020Tackling Over-Smoothing for General Graph Convolutional Networkscited by 66
  20. 2022Empowering Graph Representation Learning with Test-Time Graph Transformationcited by 66
  21. 2022Towards Robust Graph Neural Networks for Noisy Graphs with Sparse Labelscited by 63
  22. 2022Structural Entropy Guided Graph Hierarchical Poolingcited by 62
  23. 2022Assaying Out-Of-Distribution Generalization in Transfer Learningcited by 53
  24. 2021Local Augmentation for Graph Neural Networkscited by 52
  25. 2022GOOD-D: On Unsupervised Graph Out-Of-Distribution Detectioncited by 47
  26. 2023Structural Re-weighting Improves Graph Domain Adaptationcited by 45
  27. 2019DANE: Domain Adaptive Network Embeddingcited by 43
  28. 2022Non-IID Transfer Learning on Graphscited by 43
  29. 2022Tree Mover's Distance: Bridging Graph Metrics and Stability of Graph Neural Networkscited by 39
  30. 2019Independence Promoted Graph Disentangled Networkscited by 38
  31. 2023Mind the Label Shift of Augmentation-based Graph OOD Generalizationcited by 34
  32. 2021Adaptive Transfer Learning on Graph Neural Networkscited by 31
  33. 2023When to Pre-Train Graph Neural Networks? From Data Generation Perspective!cited by 17
  34. 2021A Broader Picture of Random-walk Based Graph Embeddingcited by 15
  35. 2023Explaining and Adapting Graph Conditional Shiftcited by 14

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 5; 2025: 8; 2026: 4.

17 papers.

Sort:

By area:

  1. 2024GFT: Graph Foundation Model with Transferable Tree Vocabularycited by 82
  2. 2024Towards Graph Foundation Models: Learning Generalities Across Graphs via Task-Treescited by 30
  3. 2025AgentRouter: A Knowledge-Graph-Guided LLM Router for Collaborative Multi-Agent Question Answeringcited by 29
  4. 2024Can LLMs Convert Graphs to Text-Attributed Graphs?cited by 19
  5. 2025Beyond Message Passing: Neural Graph Pattern Machinecited by 17
  6. 2026EvolveRouter: Co-Evolving Routing and Prompt for Multi-Agent Question Answeringcited by 13
  7. 2025Generative Graph Pattern Machinecited by 11
  8. 2025NG-Router: Graph-Supervised Multi-Agent Collaboration for Nutrition Question Answeringcited by 9
  9. 2024Training MLPs on Graphs without Supervisioncited by 8
  10. 2024A Pure Transformer Pretraining Framework on Text-attributed Graphscited by 6
  11. 2025Controllable Graph Generation with Diffusion Models via Inference-Time Tree Search Guidancecited by 5
  12. 2025Towards Efficient Training of Graph Neural Networks: A Multiscale Approachcited by 4
5 more
  1. 2025DELPHYNE: A Pre-Trained Model for General and Financial Time Seriescited by 3
  2. 2025A Scalable Pretraining Framework for Link Prediction with Efficient Adaptationcited by 2
  3. 2026Generalizing GNNs with Tokenized Mixture of Expertscited by 1
  4. 2026Hypergraph Pattern Machine: Compositional Tokenization for Higher-Order Interactionscited by 0
  5. 2026Refining Heuristic-Based Bitcoin Address Clustering with Graph Neural Networkscited by 0